SoX can do real time RIAA (and more) from soundcard input to soundcard output. At least it's free to try the concept. The latency does not really matter here, in any case it remains pretty low.
Last edited:
You can use it on the fly if you want. A tad more clanky than what I think you want, but vinyl is clanky anyway compared to other media.
There are choices, choices are good
Couldn't disagree more. A couple of audiophool reviews about how it improves PRAT, and digital RIAA will be the new fashion.
And you could arrange that for me? I should warn you, I have said a few things about Michael Fremer that were a bit rash.
Just pointing out it could happen. And if you really wanted to be unethical you could just plant some positive reviews on one of the countless audio review sites. I rarely read audiophool reviews nowadays, but when I used to I ran into many fishy cross-referenced online review sites.
It is S/N that worries me and was going to be my argument against. And why I never tied it, even tho I have the processing, the filters and was running my phono thru an ADC!
You don't think that the 40dB attenuation at the top end is a problem? Where would you do volume control? Before or after the ADC? One wouldn't want to attenuate the incoming analog signal too much, right?
Mini DSP should be able to do it. Jriver too if it accepts analog inputs but I don't know how great it would turn out with a soundcard or interface input unless you boosted the player's output 20dB or so before you put it into the computer.
When you convert the analog into digital you will lose all the benefits of analog audio. Instead of being infinitely variable, the audio will be chopped up to a clock and even if you use a decent analog to digital converter, if you A/B it with the analog you will be able to tell the difference. At that point you might as well play a high quality version of the digital music and it will have a way better noise floor.
You should still do it. Let us know how it sounds. Do you have a way to rip the vinyl to digital?
When you convert the analog into digital you will lose all the benefits of analog audio. Instead of being infinitely variable, the audio will be chopped up to a clock and even if you use a decent analog to digital converter, if you A/B it with the analog you will be able to tell the difference. At that point you might as well play a high quality version of the digital music and it will have a way better noise floor.
You should still do it. Let us know how it sounds. Do you have a way to rip the vinyl to digital?
Hi,
the main problem imho is that the signal delivered by the pickup is preequalized, which means a rise in amplitude response with rising frequency after the inverse RIAA.
The equalization takes place over a roundabout 40dB range.
This translates into 7Bits of the ADCs resolution.
You have also account for some headroom for clicks and pops coming off the vinyl, costing You at least 1Bit (~6dB) more, rather two.
As the signal level is also within or mostly even below the mV range alot of gain is required.
While due to the falling character a phono-pre attenuates high frequencies and as such the noise, a linear gain stage amplifies the noise fully.
At the same due to the falling openloop gain of OPAmps less and less feedback is available, leading to vastly increasing THD.
And last should the gain of the linear analog stage be configurable so that the ADCs input range is used as best as possible (not to loose more Bits here).
All in all RIAA, headroom etc. cost alot on resolution.
As the typical dynamic range is restricted to maybe 18, 19Bit (~108, 114dB) at best,You will see that not much reserve is left at low frequencies.
It should be enough for vinyl as that´s signal-to-noise ratio is probabely below 60dB (~10Bit), but it highlights the requirement for a high quality, high dynamic range ADC and highest degree of modulation.
For the digital RIAA I see three major points of merits.
The first is the interchannel balance of the equalization curve.
The second is the possibility to equalize the cartridge´s own amplitude response (think of the HF-resonance due to capacitive loading etc.).
And the third is the general possibility of signal handling and distribution in the digital domain.
jauu
Calvin
the main problem imho is that the signal delivered by the pickup is preequalized, which means a rise in amplitude response with rising frequency after the inverse RIAA.
The equalization takes place over a roundabout 40dB range.
This translates into 7Bits of the ADCs resolution.
You have also account for some headroom for clicks and pops coming off the vinyl, costing You at least 1Bit (~6dB) more, rather two.
As the signal level is also within or mostly even below the mV range alot of gain is required.
While due to the falling character a phono-pre attenuates high frequencies and as such the noise, a linear gain stage amplifies the noise fully.
At the same due to the falling openloop gain of OPAmps less and less feedback is available, leading to vastly increasing THD.
And last should the gain of the linear analog stage be configurable so that the ADCs input range is used as best as possible (not to loose more Bits here).
All in all RIAA, headroom etc. cost alot on resolution.
As the typical dynamic range is restricted to maybe 18, 19Bit (~108, 114dB) at best,You will see that not much reserve is left at low frequencies.
It should be enough for vinyl as that´s signal-to-noise ratio is probabely below 60dB (~10Bit), but it highlights the requirement for a high quality, high dynamic range ADC and highest degree of modulation.
For the digital RIAA I see three major points of merits.
The first is the interchannel balance of the equalization curve.
The second is the possibility to equalize the cartridge´s own amplitude response (think of the HF-resonance due to capacitive loading etc.).
And the third is the general possibility of signal handling and distribution in the digital domain.
jauu
Calvin
Read Channel D's white paper given at the AES. They make good points even though I think they should have also looked at crest factor. IMO the crest factor of the pre-emphasized music will help determine how much range you can squeeze out of the A/D. No mater how you slice it I doubt you would ever take the 7bit hit with any real LP/pre-amp combo.
The clicks and pops would probably be best dealt with with a clamp circuit with near instant recovery. The information is lost in any case and this is certainly doable. There are instrumentation amplifier solutions like the Forsell or Cohen circuits (based on earlier circuits but optimized for audio) that do not have "vastly" increased THD. Remember flat mV level preamps are already in the signal path, microphones.
Thanks Scott. It was mostly the click/pop levels that worried me, as they become more annoying when they clip even in the analog domain. Clipping the ADC on noise isn't going to sound nice.
If a clamp circuit is needed, are we reaching the point where using analog RIAA correction is a better idea?
If a clamp circuit is needed, are we reaching the point where using analog RIAA correction is a better idea?
Some argue for the 75usec time constant on the pre-amp and the rest digital, seems like a why bother to me. The Channel D folks have some very expensive flat pre-amps digital RIAA and a loyal following, though that's the norm in the MAC world.
I need to get up the energy to try some numerical simulations.
Just read the paper, it has some good coverage. One thing I found with ADC after RIAA is that the tone arm resonance becomes the loudest part of the signal. Channel D's paper does not show that, as they analyze down to only 20Hz.
Of course without the RIAA bass boost before the ADC, it's not a issue. It might even help.
Of course without the RIAA bass boost before the ADC, it's not a issue. It might even help.
I must admit it's the overload margin that worries me, but only because people have been saying for years that an MC phono NEEDS 40 odd dB of overload margin to sound 'good'. I think people are now realising that its recovery from transient overload that is the real issue.
As Scott points out, the instrumentation guys sorted this stuff before colour TV was widely available...
As Scott points out, the instrumentation guys sorted this stuff before colour TV was widely available...